A dynamic model to predict fat and protein fluxes and dry matter intake associated with body reserve changes in cattle.

The objective of this paper was to develop the structure and concepts of a dynamic model to simulate dry matter intake (DMI) pattern and the fluxes of fat and protein in the body reserves of cattle associated with changes in body condition score (BCS) for application within the structure of applied nutrition models. This model was developed to add the capability of evaluating the effects of factors affecting pre- and postcalving DMI, daily energy and protein balances, and changes in BCS over a reproductive cycle. Input variables are average DMI, diet metabolizable energy, and animal information (body weight, BCS, milk production, and calf birth body weight) from each diet fed over the reproductive cycle. Because the depletion and repletion of body reserves in cattle is a complex system of coordinated metabolic processes that reflect hormonal and physiological changes caused by negative or positive energy balances, the system dynamics modeling methodology was used to develop this model. The model was used to evaluate the effect of the dynamic interactions between dietary supply and animal requirements for energy and protein on the fluxes of body fat and body protein of dairy cows over the reproductive cycle and Monte Carlo simulations were used to assess the sensitivity of the parameters. The main long-term factor affecting DMI pattern was the growth of the gravid uterus causing an increase in the volume of abdominal organs and a compression of the rumen, consequentially reducing feed intake. Changes in body reserves (fat and protein) were computed based on metabolizable energy balance, assuming different efficiency of utilization coefficients for fat and protein during repletion and mobilization. The model was evaluated with data from 37 dairy cows individually fed 3 different diets over the lactation and dry periods. The model was successful in simulating the observed pattern of DMI (mean square error was 3.59, 3.97, and 3.66 for diets A, B, and C, respectively), but it tended to underpredict DMI during late lactation [around 200 to 285 d in milk (DIM)] for all diets, suggesting changes in the model structure might be needed. The predicted BCS pattern had a trend similar to the observed values. Assuming that observed BCS represents actual body fat, the model tended to overpredict observed BCS during early lactation (0.125 BCS for 0 to 120 DIM) and underpredict it during late lactation (0.06 BCS for 180 to 270 DIM). A long-term simulation (5 lactations and 4 dry periods) with diet A indicated that the cows on this diet would have a net loss of body fat if all conditions were constant.